model-interpretation

Warn

Audited by Snyk on Aug 26, 2026

Risk Level: MEDIUM
Full Analysis

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

  • Third-party content exposure detected (medium risk: 0.30). The required runtime workflow (scripts/slice_errors.py / scripts/largest_misses.py) ingests held-out prediction tables from a user-supplied --input path and then computes/prints slices and largest-miss rows based on that data, meaning outsider-authored free text can be included via the CSV/JSON/NDJSON fields.

Issues (1)

W011
MEDIUM

Third-party content exposure detected (indirect prompt injection risk).

Audit Metadata
Risk Level
MEDIUM
Analyzed
Aug 26, 2026, 06:29 PM
Issues
1
Security Audit — snyk — model-interpretation